Family Physicians in the Child Health Care Workforce: Opportunities for Collaboration in Improving the Health of Children

THE CURRENT ROLE OF FPs IN CARING FOR CHILDREN

Most children receive health care in physicians' offices, most often provided by pediatricians and FPs⁷; however, data from commercial health plans⁸and national health surveys⁹ have suggested a shift of children's physician visits away from medical generalists, specifically FPs, to pediatricians. Visit data from the National Ambulatory Medical Care Survey (NAMCS) indicate that visits by children to FPs have decreased by nearly 25% relative to a 20% rise in visits to pediatricians over the last decade (Fig 1). FPs experienced a one-third reduction in average annual children's visits per FP while visits per pediatrician remained stable (Table 1). The NAMCS data suggest that the decline in care provided by FPs to children has occurred largely in urban and suburban areas.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Trends in care of children by physicians: percentage of children's (≤18 years) visits to physician's offices. Data source: NAMCSs, 1992–2002; analysis by the Robert Graham Center, 2005.

The Medical Expenditure Panel Survey (MEPS) is another important national survey that allows profiling of children's health care with children as the unit of analysis rather than visits. Until 2002, the MEPS did not permit analyses of visits by specialty, but it is now possible to compare the profile of children's care with the profile of visits offered by NAMCS. Our MEPS analysis of children's health care found a slightly higher percentage of care provided to children in physician offices by FPs in 2002 than NAMCS (21% vs 16% of visits) and a lower percentage of care performed by general pediatricians (53% vs 60%). Despite slight differences between MEPS and NAMCS data, they collectively suggest that the trend described above is real. Despite this shift, FPs still provide care for as many as 1 in 5 children of all ages, provide more visits than pediatricians for adolescents (26% vs 24%), and are named as the usual source of care for nearly one third of children for whom a usual source of care can be named (Table 2). ^{5, 7}

THE GROWTH OF THE HEALTH CARE WORKFORCE FOR CHILDREN

The most influential factor resulting in shifting of the location of children's health care is likely the relative growth in the number of health care providers for children. Goodman et al⁵ reported a 28% absolute growth in the physician workforce between 1991 and 2001. Looking back to 1981 and focusing only on physicians who spent the majority of their time in direct patient care, we found that there was a 78% absolute increase in the total physician workforce—a 58% increase in FPs and general practitioners (GPs) and a 108% increase in pediatricians. In 2004, the number of physicians who spent the majority of their time in direct patient care in the United States and who routinely cared for children was nearly 154000. During this same period the crude birth rate declined by 11% (Table 3). As of 2004, there were ∼1600 children for every 1 pediatrician in direct patient care, or nearly 63 general pediatricians per 100000 children, nearing or exceeding some published measures of sufficiency (Table 4). ^{5, 10} There are also ∼3200 people in the United States for every 1 FP or GP in direct patient care, or 31 FPs per 100000 people, ∼88% of whom care for children.^{7, 11} If FPs represent nearly one quarter of visits by children, we conservatively estimate that there is 1 full-time equivalent (FTE) physician for every 1000 to 1200 children. This ratio is at or below most estimates of sufficiency.^{2, 5, 10}

The number of nurse practitioners (NPs) and physician assistants (PAs) caring for children is not certain, but it is likely to be at least as great as the number of pediatricians—a fact relatively unacknowledged in most workforce studies but noted recently by Goodman et al.⁵ There are no centralized NP workforce data; however, the American Nursing Credentialing Center lists 3004 nurses certified as pediatric NPs and 33288 nurses certified as family NPs (American Nursing Credentialing Center, written communication, May 25, 2005). Less than 5% of PAs work in general pediatrics and pediatric subspecialties, whereas nearly 30% work in family medicine (a total of nearly 18000).¹² Most NPs and nearly all PAs work with physicians and represent a relatively unmeasured contribution to health care for children. National health data sets do not adequately capture care provided by NPs and PAs.

In summary, the general pediatrician population grew annually at 7 times the rate of the US population between 1981 and 2004, and the FP workforce grew at nearly 5 times the rate. During that time, the general pediatrician per 100000 children ratio doubled, and the FP per 100000 children ratio grew by one third, all while the crude birth rate decreased. The growth of the US child population is expected to slow even more without any expected decline in the growth rate of the physician workforce that cares for them. Sufficiency could quickly become surplus.

DISTRIBUTION OF THE CHILD HEALTH CARE WORKFORCE

The distribution of the child health care physician workforce underscores the importance of family medicine's role in ensuring access to care for children. The growth of the pediatric workforce has largely occurred in areas of affluence and in urban or suburban areas, leading to wide variations in pediatrician-to-population ratios.^{5, 13, 14} Freed et al¹³ note that the failure of market forces to produce more level distributions despite considerable growth in the pediatric workforce runs counter to economists' predictions in the 1980s,¹⁵ and that even if the pediatric workforce continues to rise, geographic concentration will likely continue. The FP workforce tends to parallel the population in its distribution.^{16, 17} This is one reason why rural and other underserved populations depend more on FPs (Table 2). Between 1981 and 1996, rural pediatrician-to-child ratios remained well below those of urban ratios, and only in counties with populations of ≥25000 did the rural pediatrician-to-child ratio increase meaningfully.¹⁸ In 1981, the percentage of pediatric residency graduates opting for rural practice was 14.6%, but by 2001 it was just 1%.^{18, 19} Pediatrics graduates choosing to practice in rural health professional shortage areas decreased even further, and by 1996, only 123 pediatricians were practicing in rural, whole-county health professional shortage areas.¹⁸ On the basis of these findings, it has been suggested that it may not be feasible for pediatricians to practice in some counties with <25000 people. One explanation for differences in rural distribution is that family medicine draws from the entire age range in less-populated areas and is less sensitive to the relative size of the child population. Similarly, FP reimbursement draws from a broader spectrum of paying patients, and the family medicine business model is less sensitive to variations in payments for children's visits. These issues make the pediatric business model less viable in some regions, resulting in a critical need for FPs. Currently, 7% or >5 million children and adolescents live in counties with no pediatrician, but only 210000 live in counties that have neither type of physician.^20,21

THE HEALTH CARE WORKFORCE FOR UNINSURED AND UNDERSERVED CHILDREN

Pediatricians and FPs are required presently to maintain the health care safety net for children in the United States. Since 1970, the number of children covered by Medicaid doubled to more than one quarter of all children, and another 5 million are enrolled in the State Children's Health Insurance Program (SCHIP).²² The percentage of poor and near-poor children <19 years of age who were uninsured decreased by ∼25% from 1998 to 2003, and the total number of children without insurance declined from 13.2% to 10.2%.²³ Despite these improvements, >1 in 10 children remain uninsured, and an equal proportion experience unmet health care need in a given year. In 2001, >5 million children lacked health insurance for a full year, and >9 million had gaps in insurance that affected their access to health care.²⁴ In 2004, the US Census Bureau reported that 8.3 million children lacked health insurance.²⁵ In 2002, 7.3 million children had no medical home.²⁶ For the poor and near poor, particularly racial and ethnic minorities, unmet need is a more glaring disparity.²² From 1998 to 2003, there was a 15% increase in public insurance coverage such that the proportion of children covered by public insurance rose from 1 in 5 to >1 in 4, with the greatest increase among children who were near poor, whose rate more than doubled (22.5% in 1998 and 46% in 2003). Among near-poor children, those who were uninsured were more likely to have unmet medical need (35.5%) than those with public (9.4%) or private coverage (14.4%).²⁶

Children with a pediatrician as their usual source of care are significantly more likely to have private insurance, and children with no insurance or public insurance are significantly more likely to have some other usual source of care (Table 5). This insurance pattern is reversed for children with FPs as their usual source of care, but this does not reach statistical significance. This apparent reversal of insurance type and usual source of care may be explained partly by the disproportionate reliance of rural and safety-net settings and programs on FPs and suggests another important niche for FPs. What should be concerning to both specialties, however, is the fact that having any usual source of care—a medical home—is significantly associated with having some sort of health insurance, and there is a significant difference between children with private versus public insurance (Table 5).

Community health centers (CHCs) are important locations of health care for the underserved and uninsured. CHCs welcome the services of both pediatricians and FPs/GPs but depend more heavily on FPs. In 2004, 3263 FTE FPs/GPs conducted 12987000 adult and child visits in CHCs,²⁷ and 1260 FTE general pediatricians conducted 4953000 patient encounters. The National Health Service Corps is another federal program aimed at securing the safety net by supporting the placement of providers in very underserved and often rural communities. The National Health Service Corps depends on all primary care physician specialties but has consistently relied on FPs and GPs at a rate 3 to 4 times that of pediatricians.⁷ Both specialties play an important role in these safety-net programs, and health centers probably represent one of the best models of clinical cooperation between pediatricians and FPs in caring for communities. Beyond sustaining the safety net, health centers could be a laboratory for collaborative education. It is uncertain how many pediatric or family medicine residency training programs currently use CHCs for training, but they may be a suitable site for experiments in collaborative education.

OPPORTUNITY IN TRANSITIONS

FPs have a slightly larger share of the office-based visits made by adolescents than pediatricians, but combined, the 2 specialties account for just half of the visits made by those in this age group (Table 2). The decline in number of visits to pediatricians is evident in the preadolescent years, and there may be an opportunity to communicate to parents and maturing children about the need to maintain age-appropriate care, either in pediatrics or by assisting with transitions to family medicine when children and their parents feel they have “aged-out” of pediatric care. Considerably more children are surviving premature birth, cancer, congenital cardiac defect repair, and chronic disease into adulthood. The Institute of Medicine recently identified a growing problem with inadequate hand-offs and follow-up of care as these patients transition into adolescence and adulthood.²⁸ Improving transitions for all children, but particularly these special populations, could be an important organizing focus for pediatric and FP collaboration. The Future of Family Medicine report²⁹ articulates a new model of practice that may be a focused opportunity as it envisions a community population-based approach to caring for people with chronic conditions, including the use of disease registries and community-oriented primary care methods.

MOVING TO A JACOBIAN FUTURE

In 2005, the Community Pediatrics Training Initiative³ and the AAP Committee on Community Health Services⁴ called for a reengagement of the pediatrician's role in the health of communities. Satcher et al⁴ suggested that pediatricians are unable to sufficiently address health conditions precipitated or exacerbated by social, community, and environmental factors (“millennial morbidities”) and should expand their role beyond providing health care to individual patients. They specifically offered that the shrinking child-to-pediatrician ratio may be an opportunity for pediatrics to redefine itself and involve itself in more of the advocacy roles suggested by Dr Abraham Jacobi 100 years ago.³⁰ Jacobi said, “It is not enough, however, to work at the individual bedside in the hospital. In the near or dim future, the pediatrician is to sit in and control school boards, health departments, and legislatures. He is a legitimate advisor to the judge and jury, and a seat for the physician in the councils of the republic is what the people have a right to demand.” Community-oriented care and advocacy are areas of potential partnership with FPs, nurses, PAs, and others working in primary care settings.

In 2003, the AAP Task Force on the Family was asked to “help guide the development of public policy and recommend how to assist pediatricians to promote well-functioning families.”¹ The task force came to 2 “overriding” conclusions: (1) children's outcomes are strongly influenced by how well their families function, and (2) there is much that pediatricians can do to help nurture and support families. There were many additional recommendations about the training and support needed to permit pediatrician's ability to practice “family pediatrics.” It has been subsequently suggested that the task force's recommendations be included in the AAP medical home policy statement so that, “just as children have providers who know them and their needs, so too do their parents.³¹ Caring for families in the context of caring for their children is routine to family medicine. Providing care to people other than the primary patient occurs in 6% to 18% of visits to FPs, and in nearly one quarter of these visits the care is provided to parents.^{32, 33} Family medicine has adopted the AAP's concept of a medical home with attribution in reports on the Future of Family Medicine and design of new model practice.²⁹ The Future of Family Medicine report states that ”steps must be taken to ensure every American has a personal medical home that serves as the focal point through which all individuals—regardless of age, gender, race, or socioeconomic status—receive a basket of acute, chronic, and preventive medical care services.” Both pediatricians and FPs now recognize that achieving their goals requires working with families.

CONCLUSIONS

We have attempted to fill an identified gap in understanding the current role of FPs in caring for children. FPs still provide 16% to 26% of visits and are the named usual source of care for one third of the child population, but they face shrinking panels of children relative to 10 years ago. Exceptions exist in rural and safety-net sites, where family medicine's role in providing health care to children seems to be stable and vital. It remains unclear why family medicine's role is changing, and although it is likely to be a result of many factors, “saturation” of the market with more easily identified child health care providers may be a dominant factor. One concern shared by pediatricians and FPs is the failure to help adolescents, particularly those with special care needs, to effectively transition from pediatric to adult care. Another common concern is that, despite significant growth in the number of clinicians caring for children and the decline in uninsured children, 1 in 10 children still experience unmet health care needs, and 1 in 3 children without insurance have unmet health care needs.⁷ Public insurance has been a safety net for many children, but with economic downturn the number of uninsured children may rise again given state and federal efforts to reduce Medicaid costs.

Although changes in health services demand are difficult to predict, our findings demonstrate that the growth of providers of children's health care is clearly outpacing the present and expected growth of the US child population and add to concerns about a surplus of physicians caring for children. However, even with ample providers of health care for children in the US workforce to meet accepted ratios of population to provider, their distribution is skewed, leaving certain populations and settings underserved.

In keeping with recent pediatric workforce studies, we do not recommend a particular supply of pediatricians or FPs to care for children. However, in light of a constricting role in children's health care and an increasingly competitive environment for the same, family medicine is left with several options, including:

1. Relinquish clinical care for children to pediatricians and focus on working with internal medicine to meet the increased health care demands of an aging adult population.

2. Relinquish most clinical care for children and focus on preparing a segment of the FP workforce to care for children in rural and underserved sites.

3. Compete head-to-head with pediatricians, NPs, and PAs for a shrinking child health care market, working to achieve sufficient brand recognition and value to recapture market share.

4. Seriously engage pediatricians, NPs, and PAs in meaningful collaboration to build new models of training and practice that benefit from all sets of skill and compassion to provide better care in a family- and community-focused environment. This collaboration could involve joint or combined training and aggressive joint advocacy for improved services, both clinical and in the community.

Pediatrics has similar options outlined elsewhere.² A clarified and collaboratively planned role for FPs and pediatricians in caring for children is needed. Such clarification could add support to the recent statement of the AAP Committee on Pediatric Workforce, which encouraged incentives and policies that might increase the number of pediatric residents pursuing undersupplied subspecialties.⁶

The Future of Pediatric Education II report suggested that despite continued potential for competition between pediatricians and FPs, there will be new opportunities to implement cooperative models, particularly in underserved areas.¹⁰ This report also reaffirmed the recommendation that every child have a “medical home,” an “approach to providing continuous and comprehensive primary pediatric care from infancy through young adulthood.” This call from pediatrics for a medical home has been echoed by family medicine.²⁹ From a family medicine perspective, a medical home should exist for every person, serving as a reliable location at which any health concern can be received and appropriate care can be organized and integrated in a highly personal, patient-centered manner. Such care is both desired by the public and feasible to implement now, given the robust health care workforce that exists and advances in health care delivery.³⁴

Stimulated by a shrinking market for providing care to individual children, FPs and pediatricians have an opportunity to join efforts in meeting the needs of children, for many of whom their most pressing morbidity risks are framed in the context of their families and communities. To take hold, collaborative efforts must take place in medical training as well as in medical practice. Individual physicians and professional organizations will need to recognize the fundamental role of each specialty in the service of children if they are to capitalize on collective strengths.